The objective of the research is to determine if dropouts to inpatient drug and alcohol programs can be identified from psychosocial admission data and to follow dropouts and compare them with a matched group of regularly discharged Ss over six months in the community. Another objective is to test the effect of behavioral counseling for dropouts randomly assigned to treatment. Intake of Ss meeting study criteria will continue over two years, with about 210 alcohol and 320 drug patients tested at admission on social functioning, mood, symptoms, attitudes, and motivation. About 30 percent dropout is expected. Completers and dropouts will be compared on psychosocial admission data. The approximately 70 alcohol patients and 80 dropouts will be matched with an equal number of regular discharged Ss and followed in the community. The dropouts will also be randomly assigned to behavioral counseling or no planned treatment. Criterion measures of social function, mood, symptoms, attitudes, and community tenure will be repeated at one week and six months from hospital departure. In phase 1, the dependent variable will be attrition; in phase 2, it will be degree of change on the criterion measures and differences related to number of days in the community.